1897.] NEW YORK ACADEMY OF SCIENCES. 143 



The second form of permanent set embraced all those rails 

 low at the joints and centers, but high at the quarters ; this 

 form was found upon rails laid with alternate joints ; if, how- 

 ever, the traffic was heav}'^, and only a small track force em- 

 ployed, the second form would soon be converted into the first 

 form of permanent set. 



The third form of permanent set was due to a series of minute 

 waves upon the surface of the rails. Combinations of the first 

 and third, and of the second and third were frequent. 



The first and second forms of permanent set were developed 

 in the track by service, while the third form was due to improper 

 finish at the mills. 



For the rails to take a set in the track it was very evident 

 that the fibre stresses in the rails exceeded the elastic limits of 

 the steel. As the normal condition of the former iron rails in 

 the track was one in which nearly all of the rails had taken a 

 permanent set, the same features in the steel rails excited very 

 little comment among the trackmen, but among the officials the 

 matter received more serious consideration. 



It was evident that the standard of track then being main- 

 tained was too low for the increasing traffic of the roads. 



The first remedy proposed by the railroad companies was to 

 increase the labor and add more ballast to the track, and what 

 would be a proper standard of track was extensively discussed. 

 Though this was an excellent beginning, as the work increased, 

 it became evident that some more definite measure of what a 

 good track was, than merely general expressions, which differed 

 widely, would be required as a basis for further investigation, and 

 so the summing up mechanism was designed and completed for 

 the inspection of 1881. 



The number of feet of undulations per mile, as summed up by 

 the instruments on the best track of comparativel}^ new rails, was 

 about six feet per mile, while, on older rails, not in so good con- 

 dition, the amount would be from ten to twelve feet per mile. 



The general average for a number of roads was about eight 

 feet per mile. None of the tracks, particularly on the four-inch 

 rails, had what we would now term a high degree of stability, 

 for they required constant attention to keep them in what was 

 called good condition. 



In 1882 the markers were added for ejecting paint on the rails 

 where the deflections exceeded five-sixteenths of an inch. This 

 •was very convincing, and left little doubt in the minds of many 

 railroad officials that the deflections of the rails in the track 

 were greater than the}^ had anticipated. 



The rails were marked so that the officials and trackmen could 



